roc_data: Compute data to draw an ROC curve

In cboettig/earlywarning: Detection of Early Warning Signals for Catastrophic Bifurcations

Description

Grab the data for the ROC curve from the null & test distributions

Usage

  roc_data(dat, pts = 50, null_dist = NULL,
    test_dist = NULL, silent = TRUE)

Arguments

dat	a data.frame with columns "simulation", indicating "null" or "test", "value", with the value of the test statistic, and "rep" indicating the replicate number Or set dat=NULL and specify null_dist and test_dist directly, see below.
pts	(optional) number of output points desired, defaults to 50, should not be more than the number of points in each distribution
null_dist	a distribution (vector of samples) of the test statistic under the null hypothesis, if not given as a data frame
test_dist	the values under the test hypothesis

Value

a data frame with thresholds, false positive and corresponding true positive rates

cboettig/earlywarning documentation built on May 13, 2019, 2:07 p.m.